This article appeared in a journal published by Elsevier.

The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elseviers archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright

Author's personal copy

international journal of hydrogen energy 35 (2010) 585588

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/he

Technical communication

Electrolysis of LiOH for hydrogen supply

Thomas Laude a,*, Takehisa Kobayashi a, Yuzuru Sato b
a
b

Air Liquide Laboratories, 28 Wadai, Tsukuba, Ibaraki 300-4247, Japan

Tohoku University, Graduate School of Engineering, Department of Metallurgy, Sendai 980-8579, Japan

article info

abstract

Article history:

The hydrolysis of lithium hydride (LiH) is a promising technique for the supply of

Received 15 July 2009

hydrogen. LiH can be stored in sealed tanks or cartridges of w6% mass effective hydrogen

Received in revised form

capacity. However, the lack of an efficient recycling method for the byproduct lithium

6 November 2009

hydroxide (LiOH) has been a limitation to its use. Here, we propose and demonstrate

Accepted 7 November 2009

a recycling process based on the hot melt electrolysis of a LiOHLiCl eutectic mixture at 380

Available online 25 November 2009

C. The electrolysis activation potential was measured at 3.08 V. Lithium production was

observed around the cathode using differential scanning calorimetry (DSC) and cyclic
Keywords:

voltametry. Current efficiency was estimated 37.9%, and the absence of LiCl decomposition

Hydrogen

was confirmed.

Chemical storage

2009 Professor T. Nejat Veziroglu. Published by Elsevier Ltd. All rights reserved.

Electrochemistry
Molten salt
Lithium

1.

Introduction

In recent years, interest has been growing for chemicals that

reduce water and thus produce hydrogen by the simple addition of water. This concept is being considered as a hydrogen
source for hydrogen-powered vehicles (ideally, with water
recycled from the fuel cell) [1], and as a hydrogen source for
industry (ideally, with water from a pipe). Although an added
cost for the chemical processing (production or recycling) is
involved, the high density of the chemicals enables savings on
storage and distribution, compared with high pressure (HP)
solutions. Therefore, a chemical supply is appealing to isolated
regions or to situations involving space constraints.
NaBH4 has been widely studied [2], but has limitations
because of the high cost of chemical processing and complex
heterogeneous catalysis of the hydrolysis. Aluminum metal
[3] is cost competitive but the hydrolysis is slow, even when

activated. On the other hand, some alkali chemicals (Na, NaH,

Li and LiH) are strongly reductant, and therefore no hydrolysis
activation is required. LiH is particularly attractive due to its
low mass and mild pyrophoric character. LiH has been
considered as a solution for hydrogen supply [4], but an efficient LiOH recycling scheme was yet to be developed. Recycling lithium is necessary in view of its cost [5] and limited
availability.
In this study, we demonstrate a recycling approach based on
direct LiOH electrolysis, which offers the prospect of a costeffective solution. While the current efficiency is only w38% at
this stage, this technique can be expected to become cost
competitive thanks to the low activation energy of LiOH electrolysis, and the straightforward process. Alternatively,
thermal decomposition of LiOH to Li2O, followed by electrolysis
of molten Li2O (reported with carbonate-fluoride eutectic
mixture at 750900  C [6]) could be considered, but this is more

* Corresponding author
E-mail addresses: thomaslaude@uminokai.net, thomas.laude@airliquide.com (T. Laude).
0360-3199/$ see front matter 2009 Professor T. Nejat Veziroglu. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijhydene.2009.11.028

Author's personal copy

586

international journal of hydrogen energy 35 (2010) 585588

temperature demanding. Also, LiOH could be converted to LiCl

(additional step), and then electrolyzed in molten LiClKCl
(traditional method of processing lithium ore or brine [7]), but
LiCl electrolysis requires higher activation energy and higher
temperature. Other original techniques such as aqueous LiOH
electrolysis with mercury amalgam [8], or LiOH carbo-reduction [4] have also been considered.

2.
Using the LiHLiOH cycle as a hydrogen
source
For safe distribution and storage, LiH powder can be packaged
in sealed tanks or cartridges, protected from fire. Fine
powdered LiH supplied with liquid water under air can release
hydrogen with spontaneous ignition. In a dry environment,
LiH combustion is possible, with the release of caustic fumes,
although this requires a continual ignition source [9]. Lithium
compounds constitute a minor environmental issue, as they
progressively degrade to lithium carbonate on CO2 absorption
(carbonate can be recycled). Packaging in an oil slurry has also
been proposed [4], but this hinders both hydrolysis and recycling, besides reducing storage capacity.
The LiHLiOH cycle proposed here comprises three primary
reactions:
Hydrolysis, LiH H2O / LiOH H2

(1)

Molten salt electrolysis,

LiOH / Li [(1/2)-x]H2 1/2(1-x)O2 xH2O

(2)

Hydrogenation, Li 1/2H2/LiH

(3)

For hydrogen supply, the hydrolysis reaction (1) is well

known [10], and has been shown to proceed at high yield (90%,
Li metal basis) on hot vapor input [11], leaving a waste of LiOH
and water. The high affinity of LiOH for water (formation of
a mono-hydrate decomposed over 150  C [12]) increases the
waste mass but ensures that the vapor fixes on active LiH
grains rather than on the tank walls. It also ensures dry
hydrogen on output. As the LiH hydrolysis is very active, the
hydrolyzer needs no catalyst and it is suitable for high
hydrogen flow. The heat released by the active powder (DG
172.5 kJ/mol at 25  C) is expected to raise the tank temperature locally; however, this is not detrimental to the process,
as long as the heat is properly dissipated. The hydrolysis rate
is known to decrease with temperature and pressure [13], and
this is favorable for heat management. Several simple
hydrolyzer designs have been considered [11]. Fig. 1 presents
an original LiH hydrolyzer, in which hydrogen is re-circulated
to carry the water vapor to the reactor.
Theoretically the hydrogen storage capacity of LiH is as high
as 25.36% mass (H2 molar mass/LiH molar mass). However, the
practical capacity for transport (assuming an empty-to-full
exchange supply) is deduced from waste (LiOH and water)
mass. The quantity of water trapped is not precisely known, but
90% yield hydrolysis with half the LiOH hydrated, corresponds
to 6.1% mass hydrogen capacity (H2 mass/waste mass, chemical only), whereas 90% yield hydrolysis without water trapping
corresponds to 8.4% mass (waste can be dried before return).

Fig. 1 Example of a LiH hydrolyzer design. Hydrogen is recirculated to carry the water vapor from the humidifier
(a hydrophobic membrane for instance). Internal hydrogen
pressure is limited by the pre-reactor regulator.

For comparison, modern compressed storages (trailers,

bundles etc.) allow 0.50.8% mass.
Molten salt LiOH electrolysis (2) is reported here for the first
time. We use an eutectic mixture LiOHLiCl (70%30% mol 
1%) that melts at 325  C [14], and permits no Cl2 evolution
thanks to the high activation potential of LiCl electrolysis.
Lower melting temperature (LiOH melting point is 450  C)
simplifies reactor design and reduces alkaline corrosion.
Besides, it is believed that solid Li2O forms near the LiOH
melting point [15]. The technology to further hydrogenate Li to
LiH (3), and thus complete the recycle process, is available [16],
and could be continuously performed on the electrolyzer
output.
The quantity of water exhausted during the electrolysis (x)
is unknown, but is expected to be near 0 with the present setup. For comparison, the electrolysis of melted NaOH (Castner
cell) is well known (main industrial sodium production route
until 1930 [17]). Its electrode reactions are NaOH e / OH 
Na (cathode) and NaOH / Na e 1/2H2O 1/4O2 (anode).
However, water is not reported at the anode, and instead
hydrogen is produced at the cathode. Also, the current efficiency (molar ratio of Na produced to electron flow) is reportedly below 50%. This is usually explained by the migration of
the anodic H2O to the cathode and the subsequent hydrolysis
of half the Na metal produced [17]. One improvement of the
Castner design is to stop the non-ionic exchanges with an ionic
membrane placed between the electrodes. This was reported
for (H2 assisted) NaOH electrolysis [18] with 80% current efficiency. However, at present no equivalent membrane is
available for lithium-ions.

Author's personal copy

587

international journal of hydrogen energy 35 (2010) 585588

LiOH electrolysis experiment

3.1.

Experimental details

0.5

LiCl powder (Kanto Chemical 99%) was mixed with LiOH

powder (Sigma Aldrich 98%) in the proportion of LiOH 75 g
LiCl 85 g (70%30%mol  1%). The mix was placed in an
alumina crucible, and air dried for 12 h at 180  C. Temperature
was reduced to 80  C and the crucible was quickly transferred
to a quartz reactor (70  700 mm) inserted in a vertical furnace
(see Fig. 2). Three nickel electrodes (2 mm diameter) and
a thermocouple, each enclosed in an alumina tube (with inner
diameter 6 or 10 mm), were positioned in the reactor, hanging
over the mix. The junction between each alumina tube and
the enclosed electrode was sealed with epoxy or rubber. The
junction between the quartz lid and each alumina tube was
sealed with an O-ring. Reactor air was replaced with Argon,
using a rotary pump. Argon flow was set at 100 ml/min
throughout the experiment. Reactor temperature was raised
to 380  C, well over the melting point of the mixture (some
remaining moisture was vaporized and condensed around the
reactor lid). After melting, each electrode and the thermocouple were lowered 1 cm deep into the melt, measuring the
electric resistance to detect the melt surface. Electrolysis was
then started.

3.2.

Results and discussion

We performed a constant current electrolysis at 1.5 A, with

current density 1.5 A/cm2 (estimated from a tainted area on
the cathode after the experiment), close to the reported value

DSC (mcal/ sec)

3.

0
- 0.5
-1
Li melting
- 1.5
-2
100

120

140

160

180

200

Temperature (C)
Fig. 3 DSC measurement of the material surrounding
cathode. Lithium metal is identified by its melting point
signature at 184 1 8C.

of an industrial Castner electrolyzer (2 A/cm2) [17]. During the

first 30 min, we electrolyzed only the remaining moisture in
LiOHLiCl, as determined by a low inter-electrode potential
(w1.6 V), close to the water electrolysis activation potential
(1.23 V). Following this, the potential rose rapidly to the
expected potential of LiOH electrolysis, between 3.75 and 3.95
V. Temperature was stabilized at 383  C, current flowing.
Metal production rate was estimated by collecting the
material around cathode within its surrounding alumina tube,
dropping it in water and measuring the volume of (H2) gas
generated. This was done twice, at 30 min and at 46 min. The
result showed 12.5% current efficiency (molar ratio of Li
produced to electron flow) for the first 30 min (due to the
remaining moisture in LiOHLiCl), and 37.9% current efficiency in the following 16 min. Most of the current loss is
assumed to be due to inter-electrode chemical crossing,
possibly water crossing as in a Castner cell. In a separate
experiment, the metallic cathode material was analyzed by

Interelectrode current (A)

H2O

LiOH-LiCl

Fig. 2 Experimental set-up (vertical dimension reduced).

A: Quartz lid with water cooling, B: Quartz reactor,
C: Alumina crucible, D: Electrical furnace, E: Argon in/out,
F: Thermocouple in closed alumina tube, G: Nickel cathode
in open alumina tube, H: Nickel anode in open alumina tube,
I: Nickel quasi-reference electrode in open alumina tube.

2
1

Cathode reaction

Anode
reaction

-1
-2
-3

-1.5

1.5

Electrode potential (V)

Fig. 4 Cyclic voltametry measurement with potentiostat
connected to the anode, cathode and quasi-reference
electrode. The three electrodes were approximately
equidistant. The potential was measured, first between the
reference and the cathode, and then between the reference
and the anode, in two successive runs. In both runs, the
cathode-anode potential was risen until 2.5 V was reached
between the reference and the measured electrode.

Author's personal copy

588

international journal of hydrogen energy 35 (2010) 585588

differential scanning calorimetry (Mac Science DSC 3300) and

lithium metal was identified by its melting point signature at
184  1  C (Fig. 3).
The activation potential of LiOH electrolysis was deduced
as 3.08 V, from potential versus current measurement. An
estimation from the free energy of the reaction LiOH(l) / Li(l)
1/4O2 1/2H2O(g) at 700 K [19] gives 2.80 V. The difference is
expected to be equal to the LiOH ionic dissociation potential.
A cyclic voltametry measurement was done with a potentiostat measuring the potential between a quasi-reference
electrode and successively the cathode and the anode (see
Fig. 4). Anode behavior indicated a single reaction on the
potential range (expected LiOH / Li e 1/2H2O 1/4O2).
We did not observe a peak corresponding to LiCl dissociation
(LiCl / Li e 1/2Cl2), nor a peak corresponding to anode
corrosion (expected Ni 2LiOH / 2Li 2e Ni(OH)2). The
cathode showed one main reaction (expected LiOH e / OH
Li), with reverse current at 1.7 V on potential decrease (cell
working in generator mode). This indicates a (Li) metal deposition on the cathode during electrolysis. A small peak was
observed near 0.5 V in the cathodic scan, possibly linked to the
remaining water electrolysis. A minor peak was also noted
near 1.2 V, possibly resulting from a metallic impurity.
We could confirm that Cl2 was not produced on the anode
(from LiCl) by KI (potassium iodine) test. Gas from the anode
was passed through a KI solution, electrolysis being conducted at 4 V for 15 min. No color change was observed. For
comparison, we allowed pure Cl2 to flow through the same
solution, at a rate consistent with a LiCl dissociation, and
a strong color change was observed.

[2]

[3]

[4]

[5]

[6]
[7]
[8]

[9]
[10]

[11]

[12]

4.

Conclusions

In this study, the direct electrolysis of LiOH with a current

efficiency of 37.9% has been demonstrated. This offers prospect of LiOH recycling, and thus of utilizing the LiHLiOH cycle
as a hydrogen supply solution.

[13]

[14]

[15]

Acknowledgments
The electrolytic experiment was done at Tohoku University by
Takehisa Kobayashi, with technical support from Masayoshi
Hoshi.

[16]
[17]
[18]

references
[19]
[1] Satyapal S, Petrovic J, Read C, Thomas G, Ordaz G. The US
department of energys national hydrogen storage project:

progress towards meeting hydrogen-powered vehicle

requirements. Catalysis Today 2007;120:24656.
Amendola SC, Sharp-Goldman SL, Janjua MS, Kelly MT,
Petillo PJ, Binder M. An ultrasafe hydrogen generator:
aqueous, alkaline borohydride solutions and Ru catalyst.
Journal of Power Sources 2000;85:1869.
Martinez SS, Sanchez LA, Alvarez Gallegos AA, Sebastian PJ.
Coupling a PEM fuel cell and the hydrogen generation from
aluminum waste cans. International Journal of Hydrogen
Energy 2007;32:315962.
McClaine AW, Breault RW, Larsen C, Konduri R, Rolfe J,
Becker F, et al. Hydrogen transmission/storage with metal
hydride-organic slurry and advanced chemical hydride/
hydrogen for PEMFC vehicles. Proceedings of the 2000 U.S.
DOE Hydrogen Program Review, NREL/CP-57028890;2000.
Ebensperger A, Maxwell P, Moscoso C. The lithium industry:
its recent evolution and future prospects. Resources Policy
2005;30:21831.
Sintim-Damoa K, Reddy SSN, McCormick ES. Electrolytic
production of lithium metal. U.S. patent 4455202;1984.
Kipouros GJ, Sadoway DR. Toward new technologies for the
production of lithium. JOM 1998;May:246.
Cooper JF, Ebbinghaus BB, Peterman K, Weinland S,
McKenzie P. Development of bipolar cell for lithium
production. Electrochemical Society Proceedings 1995;9511:
28090.
Towndrow P. Fabrication and handling aspects of highly
reactive powders. Powder Technology 2006;174:3841.
Haertling C, Hanrahan Jr RJ, Smith R. A literature review of
reactions and kinetics of lithium hydride hydrolysis. Journal
of Nuclear Materials 2006;349:195233.
Kong VCY, Foulkes FR, Kirk DW, Hinatsu JT. Development of
hydrogen storage for fuel cell generators. I: hydrogen
generation using hydrolysis hydrides. International Journal
of Hydrogen Energy 1999;24:66575.
Cannon JC, Zhang Y. Catalytic and inhibiting effects of
lithium peroxide and hydroxide on sodium chlorate
decomposition. Industrial and Engineering Chemistry
Research 1995;34:31468.
Haertling CL, Hanrahan Jr RJ, Tesmer JR. Hydrolysis studies
of polycrystalline lithium hydride. Journal of Physical
Chemistry C 2007;111:171624.
Levin EM, McMurdie HF, Reser MK. Phase diagrams for
ceramists, 1975. Fig. 1734. The American Ceramic Society
1975;III(Suppl.):483.
Kiat JM, Boemare G, Rieu B, Aymes D. Structural evolution of
LiOH: evidence of a solid-solid transformation toward Li2O
close to the melting temperature. Solid State
Communications 1998;108(4):2415.
Alexander PP. Production of lithium hydride, U.S. patent 2
408 748;1946.
Sittig M. Sodium, its manufacture, properties and uses. New
York: Reinhold Publishing; 1956. p. 1146.
Xu J, Kelly M, Pez G, Wu Y, Sharp-Goldman S. Hydrogen
electrolysis of sodium hydroxide in molten NaOH.U.S. patent
0011662; 2004.
Chase Jr MW, Davies CA, Downey Jr JR, Frurip DJ,
McDonald RA, Syverud AN. JANAF thermochemical tables
third edition. Journal of Physical and Chemical Reference
Data 1985;14(Suppl. 1).